This invention relates to hyperspectral imaging calibration devices and methods of using the devices.
Spectroscopy, whether visible, near infrared, infrared or Raman, is an enormously powerful tool for the analysis of biomedical samples (e.g. U.S. Pat. Nos. 6,081,612; 5,784,162). The medical community, however, definitely prefers imaging methods, as exemplified by methods such as MRI and CT scanning as well as standard X-ray photography and ultrasound imaging. This is entirely understandable, as many factors need to be taken into account for a physician to make a clinical diagnosis; imaging methods potentially can provide far more information to a physician than their non-imaging counterparts. With this medical reality in mind, there has been considerable effort put into combining the power and versatility of imaging method with the specificity of spectroscopic methods.
Near-infrared (near-IR) spectroscopy and spectroscopic imaging can measure the balance between oxygen delivery and tissue oxygen utilization by monitoring the hemoglobin oxygen saturation in tissues. For in-vivo human studies, the forearm or leg has been the investigational site for many of the noninvasive near-IR studies. Non-imaging near-IR applications have examined the local response of tissue to manipulations of blood flow. Clinically, there are situations where the regional variations in oxygenation saturation are of interest. Near-IR imaging offers a means of accessing the spatial heterogeneity of the hemoglobin oxygenation saturation response to tissue perfusion.
Non-invasive monitoring of hemoglobin oxygenation exploits the differential absorption of HbO2 and Hb, along with the fact that near-IR radiation can penetrate relatively deeply into tissues. Pulse oximetry routinely supplies a noninvasive measure of arterial hemoglobin oxygenation based on the differential red-visible and near infrared absorption of Hb and HbO2. Visible/near-IR multispectral imaging permits the regional variations in tissue perfusion to be mapped. Unlike infrared thermography, hyperspectral imaging alone does not map the thermal emission of the tissues; instead, this imaging method relies on the differential absorption of light by the Hb and HbO2, resulting in differences in the wavelength dependence of the tissue reflectance depending on the hemoglobin oxygen saturation of the tissue.
Spectroscopic imaging methodologies and data are becoming increasingly common in analytical laboratories, whether it be magnetic resonance (MRI), mid-IR, Raman, fluorescence and optical microscopy, or near-IR/visible-based imaging. The volume of information contained in spectroscopic images can make standard data processing techniques cumbersome (e.g. U.S. Pat. No. 5,845,639). Furthermore, there are few techniques that can demarcate which regions of a spectroscopic image contain similar spectra without a priori knowledge of either the spectral data or the sample""s composition. The objective of analyzing spectroscopic images is not only to determine what the spectrum is at any particular pixel in the sample, but also to determine which regions of the sample contain similar spectra; i.e., what regions of the sample contain chemically related compounds. Multivariate analysis methodologies can be used to determine both the spectral and spatial characteristics of a sample within a spectroscopic imaging data set. These techniques can also be used to analyze variations in the temporal shape of a time series of images either derived for extracted from a time series of spectroscopic images.
There are few techniques that can demarcate which regions of a sample contain similar substances without a priori knowledge of the sample""s composition. Spectroscopic imaging provides the specificity of spectroscopy while at the same time relaying spatial information by providing images of the sample that convey some chemical meaning. Usually the objective in analyzing heterogeneous systems is to identify not only the components present in the system, but their spatial distribution. The true power of this technique relative to traditional imaging methods lies in its inherent multivariate nature. Spatial relationships among many parameters can be assessed simultaneously. Thus, the chemical heterogeneity or regional similarity within a sample is captured in a high dimensional representation which can be projected onto a number of meaningful low dimensional easily interpretable representations. Representations typically comprise a set of composite image each of which has a specific meaning. This is perhaps the unique element behind spectroscopic imaging.
While it is now clear that both infrared spectroscopy and spectroscopic imaging can play roles in providing medically relevant information, the raw spectral or imaging measurement seldom reveals directly the property of clinical interest. One cannot easily determine whether the tissue is cancerous, blood glucose concentrations and the adequacy of tissue perfusion.
Over the years, various methods have been devised to collect spectroscopic images. Generally in such cases, a spectroscopic imaging experiment generates data in the form of what is often called a spectral image cube, spectroscopic imaging data cube or just hypercube. This is a three dimensional array of data, consisting of two spatial dimensions (the imaging component), and one spectral dimension. The array can be thought of as an array of spatially resolved individual spectra, with every pixel in the first image consisting of an entire spectrum, or as a series of spectrally resolved images. In either representation, the 3D data cube can be treated as a single entity and it can contain enormous amounts of spatial and spectral information about the sample from which it was acquired.
A hyperspectral imaging system can acquire images at hundreds of wavelengths, as opposed to the three colors seen by humans. The added sensitivity allows such devices to accurately measure properties of the sample, such as hydration or oxygenation. However, the devices are also sensitive to many sources of variance, including both environmental conditions (such as lighting, temperature, humidity) and shading and shadowing due to object shape.
The hyperspectral image can be represented as I(x, xcex), which indicates the amount of light of wavelength xcex that falls on the imaging plane at position x. The quantity I(x, xcex) is a function of many factors that do not relate to the characteristic of the object being imaged. To accurately compute the desired object properties these factors have to be addressed. For example, the lighting conditions may involve very complex functions of both intensity and wavelength. The surface of the object may consist of various structures that cause shading changes as the normal vector to the surface changes with respect to the viewing direction.
Unfortunately, there has been no satisfactory solution to compensating the vagaries and distortions that complicate 3-dimensional imaging measurements, particularly from biological tissues or body parts. For example, U.S. Pat. No. 6,271,913 issued on Aug. 7, 2001 to Jung et al. describes a system for determining tooth color, in an effort to get around the effect of ambient light on shade color. Wunderman et al. in U.S. Pat. No. 6,122,042 issued Sep. 19, 2000 describe the use of xe2x80x9ca collection of light sources having substantially distinct wavelength envelopes and activated in a rapid sequence of distinct combinationsxe2x80x9d together with a complicated apparatus that comprises a collection of spatially distributed light detectors to optically characterize objects. The Wunderman system thus involves a great deal of hardware.
A simpler system that can detect biological problems is desired. Richards-Kortum et al. teach, in U.S. Pat. No. 6,095,982 issued on Aug. 1, 2000 the use of in vivo fluorescence measurements followed by in vitro NIR Raman measurements to detect cancerous conditions. Unfortunately, while this system requires fewer light generators and detectors, the technology uses samples that have been removed from a body and cannot accommodate various shaped body parts, particularly at a distance from the instrumentation.
Another approach that has been tried to minimize the scattering problem in the context of non-invasive glucose monitoring by optical means has been to generate sufficient signal and then limit the solid angle of illumination and detection. Block et al. in U.S. Pat. No. 6,064,065, issued May 16, 2000 describes such a system. However, this system requires particular attention to the angle of light entry and exit, and is not as suitable for biological samples that are to be imaged in three dimensions, and particularly cannot handle biological objects at a distance from the spectroscopic instrumentation.
A removable calibration device has been tried to overcome problems, as described in U.S. Pat. No. 6,045,502 issued on Apr. 4, 2000 to Epstein. However, that system cannot suitably cover a three dimensional object such as an arm or leg, and cannot create or handle imaging information from a distance. Accordingly much more remains to be done in this field.
In contrast with the above reviewed devices and methods known, the invention allows detailed spectroscopic information to be obtained from a three dimensional living object at a distance from the object. Moreover, the invention corrects the visual image information to generate higher quality images as suited for clinical analyses.
One embodiment of the invention is a hyperspectral image calibration pad, wherein the pad is form fitting to a sample to assume the shape of that sample, and has at least one surface exterior to the body part with the following optical properties: at least 95% opaque to electromagnetic radiation over at least a 50 nanometer wavelength range, preferably at least a 100 nanometer range, more preferably at least a 200, 400, 800 or more nanometer range; has a predetermined reflectance of, for example, at least 1 percent, preferably at least 2 percent, more preferably at least 5 percent, 10 percent, 25 percent or more, to wavelengths of light over said wavelength range; has a reflectance value for at least a first wavelength within said predetermined nanometer wavelength range that varies less than 10% over a desirable image region; and has a reflectance value for at least a second wavelength within said predetermined nanometer wavelength range that varies less than 10% over a desirable image region.
Another embodiment of the invention is a hyperspectral image calibration pad, wherein the pad is form fitting to a sample to assume the shape of that sample, and has at least one surface exterior to the body part with the following optical properties: at least 95% opaque to electromagnetic radiation between 400 and 500, 400 and 700, 400 and 1100, 400 and 1800, and 400 and 2400 nanometer wavelength; has a reflectance of at least 1 percent, preferably at least 2, 5, 10 or 25 percent to wavelengths of light between 400 and 1800 nm; has a reflectance value for at least a first wavelength that varies less than 10% over a desirable image region; and has a reflectance value for at least a second wavelength that varies less than 10% over a desirable image region.
Another embodiment of the invention is a method of calibrating the hyperspectral imaging of a sample, comprising the steps of: placing a disposable hyperspectral imaging calibration pad over the sample; acquiring a hyperspectral image of the calibration pad at two or more wavelengths; determining an intensity variance of a hyperspectral image that includes lighting condition information for each wavelength; and determining an intensity variance of a hyperspectral image that corrects for the shape of the object in the imaging region.
Yet another embodiment of the invention is a system for obtaining spectral information from the surface of a body part comprising: a calibrator that is form fitting to the body part to assume the shape of that part, and has at least one surface exterior to the body part that reflects light at two or more wavelengths; and a hyperspectral imaging instrument that has a light source and an imaging-quality wavelength-separation device that can image light at the two or more wavelengths to generate spectral information from the calibrator.
Yet another embodiment of the invention is a system for obtaining electromagnetic spectral information from the surface of a sample comprising: a calibrator that is form fitting to the sample to assume the shape of that sample, and has at least one surface exterior to the sample that reflects electromagnetic rays at two or more wavelengths; and a hyperspectral imaging instrument that has a light source and an imaging-quality wavelength-separation device that can image reflected electromagnetic energy at the two or more wavelengths to generate spectral information from the calibrator.
Other embodiments will be appreciated from a reading of the patent specification.